trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5029940119760479,
  "eval_steps": 4,
  "global_step": 63,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007984031936127744,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.7747,
      "step": 1
    },
    {
      "epoch": 0.031936127744510975,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.821,
      "step": 4
    },
    {
      "epoch": 0.06387225548902195,
      "grad_norm": 1.3176871932582088,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 1.8206,
      "step": 8
    },
    {
      "epoch": 0.09580838323353294,
      "grad_norm": 1.4794953169378513,
      "learning_rate": 3e-06,
      "loss": 1.8191,
      "step": 12
    },
    {
      "epoch": 0.1277445109780439,
      "grad_norm": 0.8866932750501356,
      "learning_rate": 2.8474137931034486e-06,
      "loss": 1.768,
      "step": 16
    },
    {
      "epoch": 0.1596806387225549,
      "grad_norm": 0.8287700766309047,
      "learning_rate": 2.643965517241379e-06,
      "loss": 1.8203,
      "step": 20
    },
    {
      "epoch": 0.19161676646706588,
      "grad_norm": 2.256932678210519,
      "learning_rate": 2.44051724137931e-06,
      "loss": 1.7798,
      "step": 24
    },
    {
      "epoch": 0.22355289421157684,
      "grad_norm": 1.0375119308275091,
      "learning_rate": 2.2370689655172416e-06,
      "loss": 1.7816,
      "step": 28
    },
    {
      "epoch": 0.2554890219560878,
      "grad_norm": 1.2215847471706527,
      "learning_rate": 2.0336206896551727e-06,
      "loss": 1.7728,
      "step": 32
    },
    {
      "epoch": 0.2874251497005988,
      "grad_norm": 1.1306062567864508,
      "learning_rate": 1.8301724137931035e-06,
      "loss": 1.7457,
      "step": 36
    },
    {
      "epoch": 0.3193612774451098,
      "grad_norm": 1.0863212957370068,
      "learning_rate": 1.6775862068965517e-06,
      "loss": 1.7768,
      "step": 40
    },
    {
      "epoch": 0.35129740518962077,
      "grad_norm": 0.8963887014048249,
      "learning_rate": 1.4741379310344828e-06,
      "loss": 1.7794,
      "step": 44
    },
    {
      "epoch": 0.38323353293413176,
      "grad_norm": 1.8213879073922046,
      "learning_rate": 1.321551724137931e-06,
      "loss": 1.7776,
      "step": 48
    },
    {
      "epoch": 0.4151696606786427,
      "grad_norm": 1.280443252797841,
      "learning_rate": 1.1181034482758618e-06,
      "loss": 1.7209,
      "step": 52
    },
    {
      "epoch": 0.4471057884231537,
      "grad_norm": 0.9951107450002364,
      "learning_rate": 9.146551724137933e-07,
      "loss": 1.6949,
      "step": 56
    },
    {
      "epoch": 0.47904191616766467,
      "grad_norm": 3.7028317099761736,
      "learning_rate": 7.112068965517241e-07,
      "loss": 1.7974,
      "step": 60
    },
    {
      "epoch": 0.5029940119760479,
      "step": 63,
      "total_flos": 249973539078144.0,
      "train_loss": 1.7816230607411219,
      "train_runtime": 19662.5907,
      "train_samples_per_second": 0.407,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 4,
  "max_steps": 63,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "total_flos": 249973539078144.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}